Investigating the effects of L1 proficiency and CLI : RT data from speakers of heritage L1 Turkish with dominant German L2

Abstract

This paper investigates the effects of cross-linguistic influence (CLI) in L1 Turkish of Turkish-German bilingual speakers. The study examines whether overlapping structures in the two languages result in influences of the dominant L2 German on the weaker L1 Turkish in morphosyntactic processing. Plural-marking on noun phrases was chosen for investigation since it provides an ideal test case and it constitutes partial overlap in German and Turkish. Since various definitions of CLI describe effects of this phenomenon that relate to language processing, behavioral measures are utilized in this research. The analyses of accuracy rates reveal that the two languages are clearly differentiated from each other. However, the bilingual speakers perform better with respect to the construction, which is only available in Turkish, compared to the overlapping structure between the languages. This indicates that the speakers separate the two languages from each other. However, interlanguage cue competition is at play in morphosyntactic processing in the L1 heritage language. The effects of heritage language proficiency are also examined by means of comparing high- and low-to-intermediate heritage speakers. The proficiency effects on L1 processing can be found in processing speed but not in accuracy rates. High-to-intermediate speakers do not differ from monolinguals in their processing speed, whereas the low-to-intermediate speakers perform slower than both the monolinguals and the high-to-intermediate heritage speakers. We discuss these findings within Modular Online Growth and Use of Language (MOGUL), which is a processing-based linguistic framework that accounts for the interaction of the two languages in the bilingual mind.